Download pdf - ANALYSIS OF HIGH-SPEED LITHIUM JET FLOW UNDER VACUUM … · 2019. 11. 13. · Ca=1.2, T water=297K, dissolved oxygen =5.5 mg/l Keiichi Sato et al., High Speed Observation of Periodic

KIT – University of the State of Baden-Wuerttemberg and National Research Center of the Helmholtz Association www.kit.edu

INSTITUTE FOR NEUTRON PHYSICS AND REACTOR TECHNOLOGY (INR)

ANALYSIS OF HIGH-SPEED LITHIUM JET FLOW UNDER VACUUM CONDITIONS

S. Gordeev, F. Groeschel, R. Stieglitz

3rd international Workshop onMeasuring Techniques for Liquid Metal Flows

(MTLM 2015)April 15th – 17th, 2015, Dresden, Germany

Institute for Neutron Physics and Reactor Technology (INR)-KIT

CONTENT

Int. Fusion Material Irradiation Facility (IFMIF) Development Validation strategy Recent observations – Motivation

Numerical Simulation Description of numerical model Validation of model assumption by experiments Transfer to a lithium jet

Analysis target flow Summary and Outlook

2 INR|GORDEEV, GRÖSCHEL, STIEGLITZ

IFMIF Li Target-Development


Concept of Li IFMIF (Int. Fusion Materials Irradiation Facility) Target

Goal Generation of high-energy neutron flux (simulating

fusion typical flux and neutron spectrum) Neutron generation by Deuteron (D+)–Lithium (Li)

nuclear reaction within a target (2x40MeVx125mA)10 MW target power.

Target: High speed free-surface liquid Li stream (15-20 ms−1) Upflow conditioning double-contraction nozzle Free surface flow along concave duct Ambient pressure 10−3 Pa.

Requirements Stable Li-film thickness Mechanical robustness of target system

IFMIF Li-target validation


Validation of feasibility IFMIF within Broader Approach Agreement JAEA – EU

Erection of target test loop (ELTL) including all components at prototypical scale (1:3)

Objectives ELTL demonstration of hydraulic stability

of Li target jet Li purification system in bypass

using traps (C, O2, N2).

Key parameters target Mean Li-velocity u0 = 20m/s pressure beam line p = 10−3 Pa Li-surface width w = 100mm

Observation : At low pressures accoustic noise recording @

prototypical conditions Cavitation ? (although vapour pressure Li pv=10-5Pa) initiator for present study

Optical access(beam window)

4m

uLi

safety container

Li-collector line

Quench tank

surge loop

Acoustic measurements in ELTL 1(3)


Equipment & matrix 8 acoustic emission (AE) sensors mounted on transmission bars along the downstream pipe. conventional loop instrumentation (Q, p)- monitoring test matrix (variation u0, p)

Experimental set-up

(H. Kondo, notices @ VC meeting 11th March 2015)

quenchtank

1000mm

WL (x=0)

targetassembly

x-axis along pipe bottom

origin (x=0) at weldingline (WL)

safetyvessel

© H. Kondo,2015



measurement results (sound intensity vs. pressure, u0=15m/s)

sound intensity rapidly increasing for p



Location of noise

u0=12.1 m/s

u0=15.1 m/s

flow directionWL (x=0)

x (=180°-bottom) [mm]

shift of epicenter in x-direction for rising u0

circumferrentialangle [°]

circumferrentialangle [°]

Jet trajectory motion

WL, x=0

tancos2

222

0

xxu

gy

u0=5m/s

u0=12.1m/s

u0=15.1m/s u0=17.1m/she

ight

[m]

length [m]

Numerical simulation- model description


Basis Multi-phase approach Formulation Phase interaction control by means of Volume Of Fluid (VOF) technique Implementation of cavitation model (Li-liquid (l), Li-gas (g))

Seed-based mass transfer modeln0-density of seeds (1012 m3), R-bubble radius (5.10-7m) set to default values of water (absence of Li-data)

Inertia controlled cavitation bubble growth model ( Rayleigh-Plasset equation) psat -saturation pressure, p0 - pressure of the surrounding liquid, - liquid density

VOF-free-surface model (Li-liquid (l)/ Ar-gas) Surface tension modelled by continuum surface force (CSF) technique (super-position of

normal and tangential force variation along interface) n / t – normal/tangential, unity vectors - surface tension,

K - interface curvature, wall boundary conditions: capillary effects and contact angle gravity as volumetric force fg=g

lg

gg VV

V

3

0

30

3/413/4

RnRn

g

l

sat ppDtDR

0

2

32

tn fff ,, nKf n , ttf t

,

Numerical simulation- model validation

9 S. Gordeev

Experimental conditions:Ca=1.2, Twater=297K, dissolved oxygen =5.5 mg/lKeiichi Sato et al., High Speed Observation of Periodic Cavity Behavior in a Convergent-Divergent Nozzle for Cavitating Water Jet, J. of Flow Control, Meas. & Vis., 2013

Cavitation number Ca 202/1 upp

Ca g

Cavitation nozzle

p1

p2

model qualification by water experiment (literature) submerged water jet in water pool

Expectation: u0>ucrit cavitation in jet flow


CFD

black regionsindicate

cavitation clouds

geometry model

Modeling: commercial

Star-CCM+ gas dissolution

(O2=5.5 mg/l)O2 ,H2O-vapour

– ideal gas 7.105cells

massflow inlet

wall

u0

pressureoutlet

experimental observation periodic behaviour of cavitation cloud in the nozzle

Exp.

time t [ms]

agreement in shape and temporal behaviour


Quantitative stochastic analysisNumerical simulation- model validation

40mm

volume fraction liquid

temporal progression (Sim.)

390 Hz

CFD

f [hz]

Pressure fluctuationscaused by theplunger pump

FFT of gray level change in cavity length

FFT analysis for calculated fluctuations of volume fraction of water gaseous phase

frequency of cavitylength (x=40mm)

Exp.

f [Hz]

348 Hz

reasonable cavitation model

u0



qualification of gravitation, surface tension and contact angle model inclined water jet impinging on plane vertical plate (water/air) *

g

jet diameter di=3mm=45°,

* T.Wang et al., Chemical Engineering Science 102 (2013)

u0

Film flow behind target (50 g/s)

draining film

hydraulicjump

excellent shape agreementl[

mm

]

Film shape (Z,R) as function mass flow ( or u0)

R exp.

Z exp.

R sim.

Z sim.

[kg/s] perfect agreement in Z max. deviations 7% in Rmodel conceived to adequate to

depict free surface with caviation

Z

2R

Numerical simulation - transfer to a Lithium jet


Modeling in commerical code (Star-CCM+) Realizable k-ε turbulence model Volume of Fluid (VOF) multi-phase model cavitation, surface tension, gravityComputational domain and grid 3d using symmetry conditions, 1.5x106 (u0=5m/s) up to 7.2x106 cells (u0=15m/s)

impingement domain

pressure outlet

Initial conditions: u0= 5, 15 m/s pg= 103, 10-3 Pa Li=500kg/m3const. Li=0.41N/m Ar, Li vapour– ideal gasBoundary conditions: u|wall=0 hydraulically smooth walls contact angle 60° t=10-5s

u0

Analysis –target flow (ref. operating conds. 1/4)

14

Iso-surface of liquid-lithium phase VF=0.7 (comp. time 1.5 s)

Impingementpoint x=15 mm

g

Li filmflow

Conditions: u0=15 m/s, p=10-3 Pa

WL, x=0

u0

u0 Free surface velocity in jet impinging region

reversed flow

momentum initiated upstream motion of droplets


15

Jet flow – Lithium(l) iso-surface VF=0.7Lithium gas/liquid mixture iso-surface Li(g) 5%


Lithium vapour mainly upstream impingement position



WL, x=0

Liquid Phase Gas Phase

Interaction of impinging jet with reversed flow intense small scaled bubble generation mixing of liquid and gaseous Lithium ensuing transfer of gaseous lithium into the impinging area bubble collapse (cavitation)


g

u0

Volume fraction of Li(l)

u0

Volume fraction of Li(g)

g

WL, x=0



Observation Measured length of AE area (100-120 mm)

corresponds to area with high concentration of Li(g) on the pipe wall Deviation from measured position of AE

epicenter and calculation is x 50 mm.

Conditions: u0=15 m/s, p=10-3 Pa Does observed location coincide

with experimental observation ?

(H. Kondo, notices @ VC meeting 11th March 2015)

x (=180°-bottom) [mm]

Volume fraction of Li(g) on the pipe wall

WL x=0

estimatedlocation of AE

WL x=0Max 4.5x104 Pa

Static pressure on the pipe wall

impingement point x=15 mm

© H. Kondo,2015

Reasons for uncertainties in computed position of jet impingement position


Potential sources miscellaneous flow reading initial velocity (1m/s) x 7-8 mm Improper jet cross-section shape negl. impact due to momentum governed problem Mismatch exp. geometry model large impact (!!!) Manufacturing –mismatches: Examples

x 80mm

1°

1° misalignment from pipe axis Variation of normal distance y from the pipe wall to the jet inlet 10mm

x 40mm

Analysis –target flow

Analysis –target flow


Liquid Phase

Gas Phase

Li (l) jet iso-surface VOF=0.7

absence of Li(g) fraction (experiment @ p=30kPa) Potential sources dependence of cavitation model on the mesh resolution absence of reliable data on bubble seed density, initial bubble

radius for alkali metals gases dissolved in the experiment ???

Conditions: u0=15 m/s, p=103 Pa

Conclusions &Ooutlook

No occurrence of cavitation in Li jet bulk during at nominal conditions at nominal conditions in simulation (10-3 Pa, 15 m/s).

wall impingement partial backward flow droplet formation free surface increase Li vapor production, enough to lead to significant vapor fraction amount. Li vapor captured and reintroduced in the main flow. recovery of static pressure by transport bubble collapse cavitation

Epicenter of cavitation can be predicted with accuracy of 50 mm. Deviations to experiment can be attributed to several sources (mainly geometric imperfections)

Exp. observed cavitation even at 30 kPa and u0= 15 m/s cannot be depicted numerically numerical sensitivity study underway , but likely modeling parameters (seed properties, scaling of bubble growth rate) requires

complementary model experiments.
